Enhancing the Smartphone User Experience
Always at hand, smartphones allow us to remain ever connected. AI (Artificial Intelligence) is making smartphones even smarter—from improving video/image capture to facial ID, AI in smartphones is increasing dramatically as new, compelling user experience models emerge. But not all AI is created equal, and implementing the wrong Neural Processing Unit (NPU) in an SoC/ASIC can lead to a poor user experience, including reduced battery life or slow performance.
Smartphone AI Use Cases
Today’s smartphones deploy edge AI for facial ID, object recognition, video and image enhancement, always-sensing, and immersive virtual reality. With multiple high-resolution cameras and popular social media apps driving increased video streaming, AI is the ideal vehicle to improve video quality. Smartphone makers are deploying neural networks in the traditional ISP (Image Signal Processor) pipeline to control image quality functions like demosaicing, denoising, deblurring, and super-resolution, as well as in security and access features like facial ID. Eventually, neural networks may replace the entire ISP pipeline.
Reducing AI Power Consumption in Smartphones
One of the crucial performance metrics for a smartphone is battery life. NPUs can consume significant power—that is, if the wrong NPU is chosen. Comparing the power efficiency of NPUs can be complicated. Ours isn’t—Expedera’s Origin™ IP averages a market-leading 18 TOPS/W. Origin has repeatedly been cited as the most power-efficient NPU available by third parties and customers alike.
Always-sensing NPU Support
Like always-listening audio applications, always-sensing cameras enable a more natural and seamless user experience. However, camera data has quality, richness, and privacy concerns which require specialized AI processing. OEMs are turning to specialized “LittleNPU” AI processors to process always-sensing data. Expedera’s E1 family has been optimized to process the low-power, high-quality neural networks used by leading OEMs in always-sensing applications while maintaining low power (often as low as 10-20mW) and keeping all camera data within the LittleNPU subsystem, working hand in hand with device security implementations to safeguard user data.
Optimizing AI for Smartphones
While many general-purpose AI processors exist, a one-size-fits-all solution is rarely the most efficient. General-purpose AI processors are often much larger than needed for smartphones and will consume more power than necessary. Expedera’s Origin E2 IP cores are precisely optimized for smartphone use cases. The E2 family of NPUs reduces power consumption for all smartphone use cases, increasing battery life and improving the user experience. No other AI processor can offer the performance, power, and size advantages Expedera’s Origin IP offers smartphone ASIC and ISP developers.
to our News
Sign up today and receive helpful
resources delivered directly
to your inbox.